Claude Haiku 4.5 pricing, context window and limits
Claude Haiku 4.5 costs $1 per million input tokens and $5 per million output tokens at Anthropic list price. It accepts up to 200,000 tokens of context and returns up to 64,000 tokens.
- Model id
claude-haiku-4-5-20251001 - Snapshot
What does Claude Haiku 4.5 cost?
| Input | $1 |
|---|---|
| Output | $5 |
| Cached input (read) | $0.1 |
| Cache write | $1.25 |
US dollars per million tokens. A dash means no price is published.
How much can Claude Haiku 4.5 read and write?
| Context window | 200,000 |
|---|---|
| Max output | 64,000 |
What can Claude Haiku 4.5 do?
| Vision (image input) | Yes |
|---|---|
| Tool calling | Yes |
| Structured output | Yes |
| Reasoning | Yes |
| Input | text, image, pdf |
| Output | text |
What does a request to Claude Haiku 4.5 cost?
| A typical request: 1,000 tokens in, 500 out | $0.0035 |
|---|---|
| A long request: 10,000 tokens in, 1,000 out | $0.015 |
| 1,000 typical requests | $3.5 |
| 1 million tokens in and 1 million out | $6 |
Input read from the provider's prompt cache costs $0.1 per million tokens instead of $1, 90% less. A request that repeats a 1,000-token prompt from the cache costs $0.0026 instead of $0.0035.
Among 17 Anthropic models with a published price, Claude Haiku 4.5 is the 2nd cheapest by input price. Its context window is the 12th largest of 17. List prices only; discounts, batch prices and prices above 200K tokens are not shown.
How do I call Claude Haiku 4.5 through DVARA?
Configure the anthropic provider in the DVARA LLM Gateway and send the model id claude-haiku-4-5-20251001. DVARA, the AI gateway, applies your policies, audit and cost attribution to the call. See LLM cost attribution and budgets.
In DVARA, if you haven't set your own price for a model, its calls are costed at the price in the catalogue that ships with your DVARA version. A price you set in Flightdeck always wins. You can switch catalogue prices off. Then a model with no price you set gets no cost record: its calls count against no budget and appear in no report, and the gateway records the model as unpriced. Every cost record says which price it used.
Which other Anthropic models are there?
- Claude Haiku 4.5 (latest)$1 in · $5 out · 200K context
- Claude Haiku 5.5$0.1 in · $0.5 out · 1M context
- Claude Fable 5.1$10 in · $50 out · 1M context
- Claude Opus 4.5 (latest)$5 in · $25 out · 200K context
- Claude Fable 5$10 in · $50 out · 1M context
- Claude Opus 4.5$5 in · $25 out · 200K context
- Claude Opus 4.6$5 in · $25 out · 1M context
- Claude Opus 4.7$5 in · $25 out · 1M context
- Claude Opus 4.8$5 in · $25 out · 1M context
- Claude Opus 5$5 in · $25 out · 1M context
- Claude Opus 5.5$4 in · $20 out · 1M context
- Claude Sonnet 4.5 (latest)$3 in · $15 out · 200K context
Which models from other providers are similar to Claude Haiku 4.5?
These are the closest in context window and input price, one per provider.
- Claude Haiku 4.5Amazon Bedrock · $1 in · $5 out · 200K context
- GLM-5Zhipu GLM · $1 in · $3.2 out · 205K context
- o3-miniOpenAI · $1.1 in · $4.4 out · 200K context
- Grok Build 0.1xAI Grok · $1 in · $2 out · 256K context
- Kimi K2.6Moonshot AI · $0.95 in · $4 out · 262K context